← Back to Skills Marketplace
2810
Downloads
0
Stars
91
Active Installs
1
Versions
Install in OpenClaw
/install self-improving-1-2-16
Description
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
Usage Guidance
Install only if you want the agent to keep long-term local memory of corrections, preferences, and work patterns. Review edits to AGENTS.md, SOUL.md, and HEARTBEAT.md, inspect ~/self-improving/ periodically, avoid storing sensitive information there, and treat “forget everything” carefully because the artifacts may export memory before deleting it.
Capability Assessment
Purpose & Capability
The local memory, correction logging, self-reflection, and heartbeat maintenance match the stated self-improving-agent purpose.
Instruction Scope
Instructions tell the agent to log corrections and write learned patterns automatically, and the delete-all flow is inconsistent about whether memory export is optional.
Install Mechanism
No executable scripts or required binaries are included; setup creates files under ~/self-improving/ and may install a separate Proactivity skill only after user agreement.
Credentials
The skill writes durable home-directory memory and edits AGENTS.md, SOUL.md, and HEARTBEAT.md steering, which is purpose-aligned but affects future sessions and should be explicitly reviewed.
Persistence & Privilege
Cross-session memory and heartbeat state are central to the skill, but the wipe path can create an exported copy of learned data without a clearly specified path or opt-out in all instructions.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install self-improving-1-2-16 - After installation, invoke the skill by name or use
/self-improving-1-2-16 - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Self-Improving + Proactive Agent version 1.2.16 clarifies workflow and safety rules:
- Clarifies setup flow for proactive follow-through and safer installation behavior.
- Expanded documentation on memory architecture, usage triggers, and self-reflection routines.
- Improved guidance for tiered memory promotion/demotion and conflict resolution.
- Explicit rules for transparency, compaction, and namespace isolation.
- Added table-based quick references, traps to avoid, and clear boundaries on skill scope and data storage.
Metadata
Frequently Asked Questions
What is Self Improving 1.2.16?
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when... It is an AI Agent Skill for Claude Code / OpenClaw, with 2810 downloads so far.
How do I install Self Improving 1.2.16?
Run "/install self-improving-1-2-16" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Self Improving 1.2.16 free?
Yes, Self Improving 1.2.16 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Self Improving 1.2.16 support?
Self Improving 1.2.16 is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).
Who created Self Improving 1.2.16?
It is built and maintained by Taron M. (@taron-ai); the current version is v1.0.0.
More Skills